. . . . . .

Tidal Stream Turbine Modelling using High Performance ComputingI.Afgan, J. McNaughton, S. Rolfo, D. Apsley, T. Stallard & P. Stansby

Modelling and Simulation Centre, School of MACE, The University of Manchester, United Kingdom.

.Introduction..

......

With the world supply of oil and gas fast depleting, significant interest is growing towards renewable energy. One such major source is the untapped energy coming fromoceans in the form of tidal or wave energy. However, there is limited understanding of how the wakes of marine hydro kinetic devices such as tidal stream turbines (TST ) areaffected by the ambient turbulence of the tidal flows. Full scale experimental or field measurements of such devices are both costly and extremely difficult. To complementfield trials, detailed insight into the flow physics of turbulent flow, turbine loading and wake dynamics field trials can also be obtained by Computational Fluid Dynamics.Such complex coupled problems require High Performance Computing. Our research analyses the suitability of employing massively parallel CFD for horizontal axis TSTload predictions and wake analysis to enable efficient design optimization.

.Project Breakdown/Milestones..

......

Implementation of a new sliding mesh technique in Electricite de France (EDF ) open sourceCFD solver (Code Saturne) to enable simulations of rotors within a channel.

Benchmarking a lab scale Tidal Stream Turbine (TST ) model for optimized codemodifications for High Performance Computing (HPC ).

Implementation of moving free surface boundary in Code Saturne to model wave motion,thereby, coupling rotating TST turbines with waves.

Benchmarking and design optimization for Tidal Generation Limited (TGL) 1MW TST, withactual field measurements.

.Modelling..

......

(a) (b)

Figure 1: (a) Full TST CFD domain (b) View of the surface mesh.

(a) (b) (c) (d)

Figure 2: Sliding mesh procedure optimization (a-c) Various communication methodologies between processors (d)

Speedup with different methodologies.

Complete horizontal axis laboratory scale TST with 3 blades, nacelle and a mast (see Fig. 1).

CFD simulations performed with both Reynolds Average Navier-Stokes (RANS) and LargeEddy Simulation (LES) using Code Saturne.Numerical simulations performed on EDF Blue-Gene P super-computer with fine simulationscosting about 4.4 million CPU hours for each run (approximately 4096 cores for 45 days).

Optimization of Code Saturne’s modified subroutines for HPC scalability (see Fig. 2)..Results and Discussion..

......

Large vortex shedding behind the vertical support with strong flow meandering (see Fig. 3).

Three-dimensional helical wrapping of the flow around surface of nacelle and extendingdownstream beyond the mast (see Fig. 4).

Shedding from the mast becomes inclined as it propagates downstream and is broken down bythe large scale structures of the tip vortices.

Mean blade loadings, thrust (CT ) and power (CP) coefficient from LES are within 3% ofmeasurements (see Fig. 5).

Figure 3: Instantaneous velocity and pressure field.

.

......

Figure 4: 2nd invariant of the velocity gradient tensor (Iso-Q).

(a) (b)

Figure 5: Mean performance coefficient comparison at various tip speed ratios (TSR) (a) Mean Thrust[CT =

(FX/

(0.5ρU2

0A))]

(b) Mean Power[CP =

((TQΩR) /

(0.5ρU3

0A))]

.

Pressure and velocity fields with different wave speeds (ratio of a characteristic velocity to agravitational wave velocity, Fr = (v/c)) are shown in Figs. 6 & 7.

The velocity on the upper part of the domain is no longer uniform, whereas underneath awave, the turbine is subjected to pressure waves.

(a) (b)

Figure 6: TGL 1MW TST with waves (Fr = 0.09) (a) Pressure field (b) Velocity field.

(a) (b)

Figure 7: TGL 1MW TST with waves (Fr = 0.12) (a) Pressure field (b) Velocity field.

.Conclusions..

......

A new formulation has been developed for implementing sliding meshes in Code Saturne in order

to model tidal stream turbines with and without waves. A lab scale model was tested with HPC

optimization, resulting in CFD predictions which were within 3% of the experimental data. The

latter half of the study was based on wave modelling on top of TST. The research also encompasses

the numerical issues related to the modelling of tidal stream turbines. To the authors’ knowledge,

this is the only LES of a TST with blade geometry resolved..Future work..

......

Assess the accuracy of unsteady loads predicted by the present methodology by developing incident

flow with large-scale turbulence representative of the EMEC test-site and comparing loads to full-

scale measurements..Acknowledgments..

......

This research was performed as part of the Reliable Data Acquisition Platform for Tidal (ReDAPT ) project commissioned and funded by the Energy Technologies Institute (ETI ). The authors are highly

grateful to EDF for additional funding and access to its HPC facilities..Publications..

......

I. Afgan, J. McNaughton, D. Apsley, S. Rolfo, T. Stallard & P. Stansby. Large-Eddy Simulation of a 3-bladed Horizontal Axis TST : Comparisons to RANS and Experiments. In Proc. Turbulence, Heat and Mass Transfer 7. Sicily, Italy, 2012.

J. McNaughton, S. Rolfo, D. Apsley, I. Afgan, P. Stansby & T. Stallard. CFD Prediction of Turbulent Flow on a Laboratory Scale TST using RANS modelling. In 1st Asian Wave and Tidal Conference Series. Jeju, South Korea. 2012.

I. Afgan, J. McNaughton, S. Rolfo, D. Apsley, T. Stallard & P. Stansby. Turbulent Flow and Loading on a TST by LES and RANS. Under review Int. J. Heat and Fluid Flow.

J. McNaughton, I. Afgan, D. Apsley, S. Rolfo, T. Stallard & P. Stansby. A Robust Sliding-Mesh Interface Procedure and its Application to the CFD Simulation of a TST. Under review Int. J. Numerical Method Fluids..References..

......

F. Archambeau, N. Mechitoua, and M. Sakiz (2004). Code Saturne: A Finite Volume Code for the Computation of Turbulent Incompressible Flows - Industrial Applications. Int. J. Finite, Vol. 1 (1).

A. S. Bahaj, A. F. Molland, J. R. Chaplin, and W. M. J. Batten (2007). Power and Thrust Measurements of Marine Current Turbines under Various Hydrodynamic Flow Conditions in a Cavitation Tunnel and a Towing Tank. Renewable

Energy, Vol 32 (3), 407426.

R. McSherry, J. Grimwade, I. Jones, S. Mathias, A. Wells and A. Mateus. 3D CFD Modelling of Tidal Turbine Performance with Validation against Laboratory Experiments. In Proc. 9th European Wave and Tidal Energy Conference,

University of Southampton, UK, 2011.

A. Jimenez, A. Crespo, E. Migoya & J. Garcia. Advances in Large-Eddy Simulation of a Wind Turbine Wake. In Journal of Physics Conference Series, 75(1), 012041 (2007).

Regional Lecture and Dinner, 5th March 2013, University of Manchester, UK